Matches in SemOpenAlex for { <https://semopenalex.org/work/W4292945972> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4292945972 abstract "Current methods for regularization in machine learning require quite specific model assumptions (e.g. a kernel shape) that are not derived from prior knowledge about the application, but must be imposed merely to make the method work. We show in this paper that regularization can indeed be achieved by assuming nothing but invariance principles (w.r.t. scaling, translation, and rotation of input and output space) and the degree of differentiability of the true function. Concretely, we derive a novel (non-Gaussian) stochastic process from the above minimal assumptions, and we present a corresponding Bayesian inference method for regression. The mean posterior turns out to be a polyharmonic spline, and the posterior process is a mixture of t-processes. Compared with Gaussian process regression, the proposed method shows equal performance and has the advantages of being (i) less arbitrary (no choice of kernel) (ii) potentially faster (no kernel parameter optimization), and (iii) having better extrapolation behavior. We believe that the proposed theory has central importance for the conceptual foundations of regularization and machine learning and also has great potential to enable practical advances in ML areas beyond regression." @default.
- W4292945972 created "2022-08-24" @default.
- W4292945972 creator A5014548505 @default.
- W4292945972 date "2022-08-22" @default.
- W4292945972 modified "2023-09-27" @default.
- W4292945972 title "Scale invariant process regression: Towards Bayesian ML with minimal assumptions" @default.
- W4292945972 doi "https://doi.org/10.48550/arxiv.2208.10461" @default.
- W4292945972 hasPublicationYear "2022" @default.
- W4292945972 type Work @default.
- W4292945972 citedByCount "0" @default.
- W4292945972 crossrefType "posted-content" @default.
- W4292945972 hasAuthorship W4292945972A5014548505 @default.
- W4292945972 hasBestOaLocation W42929459721 @default.
- W4292945972 hasConcept C105795698 @default.
- W4292945972 hasConcept C118615104 @default.
- W4292945972 hasConcept C121332964 @default.
- W4292945972 hasConcept C126255220 @default.
- W4292945972 hasConcept C132459708 @default.
- W4292945972 hasConcept C154945302 @default.
- W4292945972 hasConcept C163716315 @default.
- W4292945972 hasConcept C190470478 @default.
- W4292945972 hasConcept C2524010 @default.
- W4292945972 hasConcept C2776135515 @default.
- W4292945972 hasConcept C28826006 @default.
- W4292945972 hasConcept C33923547 @default.
- W4292945972 hasConcept C37914503 @default.
- W4292945972 hasConcept C41008148 @default.
- W4292945972 hasConcept C61326573 @default.
- W4292945972 hasConcept C62520636 @default.
- W4292945972 hasConcept C74193536 @default.
- W4292945972 hasConcept C81692654 @default.
- W4292945972 hasConcept C83546350 @default.
- W4292945972 hasConcept C99844830 @default.
- W4292945972 hasConceptScore W4292945972C105795698 @default.
- W4292945972 hasConceptScore W4292945972C118615104 @default.
- W4292945972 hasConceptScore W4292945972C121332964 @default.
- W4292945972 hasConceptScore W4292945972C126255220 @default.
- W4292945972 hasConceptScore W4292945972C132459708 @default.
- W4292945972 hasConceptScore W4292945972C154945302 @default.
- W4292945972 hasConceptScore W4292945972C163716315 @default.
- W4292945972 hasConceptScore W4292945972C190470478 @default.
- W4292945972 hasConceptScore W4292945972C2524010 @default.
- W4292945972 hasConceptScore W4292945972C2776135515 @default.
- W4292945972 hasConceptScore W4292945972C28826006 @default.
- W4292945972 hasConceptScore W4292945972C33923547 @default.
- W4292945972 hasConceptScore W4292945972C37914503 @default.
- W4292945972 hasConceptScore W4292945972C41008148 @default.
- W4292945972 hasConceptScore W4292945972C61326573 @default.
- W4292945972 hasConceptScore W4292945972C62520636 @default.
- W4292945972 hasConceptScore W4292945972C74193536 @default.
- W4292945972 hasConceptScore W4292945972C81692654 @default.
- W4292945972 hasConceptScore W4292945972C83546350 @default.
- W4292945972 hasConceptScore W4292945972C99844830 @default.
- W4292945972 hasLocation W42929459721 @default.
- W4292945972 hasLocation W42929459722 @default.
- W4292945972 hasOpenAccess W4292945972 @default.
- W4292945972 hasPrimaryLocation W42929459721 @default.
- W4292945972 hasRelatedWork W1741715688 @default.
- W4292945972 hasRelatedWork W2083795088 @default.
- W4292945972 hasRelatedWork W2376872241 @default.
- W4292945972 hasRelatedWork W2388787028 @default.
- W4292945972 hasRelatedWork W3185162181 @default.
- W4292945972 hasRelatedWork W4226221006 @default.
- W4292945972 hasRelatedWork W4287073795 @default.
- W4292945972 hasRelatedWork W4289711124 @default.
- W4292945972 hasRelatedWork W4366308437 @default.
- W4292945972 hasRelatedWork W4384129965 @default.
- W4292945972 isParatext "false" @default.
- W4292945972 isRetracted "false" @default.
- W4292945972 workType "article" @default.